This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. Protein aggregation is a fundamental challenge in biology and industry for its role in several neurodegenerative diseases such as Alzheimers, Parkinson, Huntington to name a few. It is believed that protein aggregates are directly involved in the etiology of disease. Understanding the structure and stability of protein aggregates is thus of utmost importance from both biochemistry and therapeutics point of view. Because of their large size and low solubility existing methods for high-resolution structure determination have not been successful producing accurate models. However, techniques such as spin labeling and fluoresence give us important indirect information about various candidate structures of these aggregates. Here we propose a molecular dynamics based simulation technique to gain further insight about such candidate structures obtained from experiments. We will start with some of these possible aggregate topologies indicated by experiments and perform long molecular dynamics simulation to study the relative stability of these structures. This will be performed in general for tau protein aggregates for its relevance in Alzheimers disease. We will use Amber force field to carry out this ab initio study which will require high performance computing. Thus we believe our simulation approach will be complementary to experimental results obtained in lab and will provide us with better idea about the stability of protein aggregates.